Resources

Novee research shows Anthropic, Google and OpenAI agent pipelines leak secrets even after safety checks pass. The failure is composed trust across stages.

• ai-security
• llm-security
• agentic-ai
• prompt-injection
• red-teaming

OpenAI models escaped a sandbox and breached Hugging Face during evaluation. Full timeline, techniques used, and key lessons for AI security architects.

• ai-security
• llm-security
• agentic-ai
• red-teaming
• sandbox-escape

GhostApproval: symlink flaw in AI coding tools (Claude Code, Cursor, Amazon Q) lets malicious repos escape sandboxes and access sensitive files.

• ai-security
• prompt-injection
• agentic-ai
• llm-security
• devsecops
• red-teaming
• sandbox

GitLost shows how one public GitHub Issue can leak private repo data via prompt injection in Agentic Workflows. Key lessons for securing agentic systems.

• ai-security
• prompt-injection
• agentic-ai
• llm-security
• red-teaming
• github

Mapping OWASP Top 10 for Agentic App risks to 2026 architecture patterns, incorporating GPT-5.6 over-agency findings for practical stack-level defenses.

• ai-security
• agentic-ai
• owasp
• llm-security
• workload-identity
• governance

GuardFall shows how classic Bash tricks bypass AI coding agents' safety filters. Learn why regex guards fail and how to build resilient tool-use layers.

• ai-security
• agentic-ai
• llm-security
• red-teaming
• devsecops

Dual-firewall and adaptive out-of-band defense patterns for securing agentic and RAG systems against prompt injection.

• ai-security
• prompt-injection
• agentic-ai
• rag-security
• llm-agents
• multi-agent-systems

Most organizations know AI governance matters but few know where to begin. Here are the first three questions every CISO should answer before scaling AI.

• ai-governance
• ciso
• compliance
• getting-started